Epidemics of hepatitis E, an enterically transmitted non-A/non-B hepatitis, have been reported in Asia, Africa and Central America (Balayan, M. S. (1987), Soviet Medical Reviews, Section E, Virology Reviews, Zhdanov, O-V. M. (ed), Chur, Switzerland: Harwood Academic Publishers, vol. 2, 235-261; Purcell, R. G., et al. (1988) in Zuckerman, A. J. (ed), "Viral Hepatitis and Liver Disease", New York: Alan R. Liss, 131-137; Bradley, D. W. (1990), British Medical Bulletin, 46:442-461; Ticehurst, J. R. (1991) in Hollinger, F. B., Lemon, S. M., Margolis, H. S. (eds): "Viral Hepatitis and Liver Disease", Williams and Wilkins, Baltimore, 501-513). Cases of sporadic hepatitis, presumed to be hepatitis E, account for up to 90% of reported hepatitis in countries where hepatitis E virus (HEV) is endemic. The need for development of a serological test for the detection of anti-HEV antibodies in the sera of infected individuals is widely recognized in the field, but the very low concentration of HEV excreted from infected humans or animals made it impossible to use such HEV as the source of antigen for serological tests and although limited success was reported in propagation of HEV in cell culture (Huang, R. T. et al. (1992), J. Gen. Virol., 73:1143-1148), cell culture is currently too inefficient to produce the amounts of antigen required for serological tests.
Recently, major efforts worldwide to identify viral genomic sequences associated with hepatitis E have resulted in the cloning of the genomes of a limited number of strains of HEV (Tam, A. W. et al. (1991), Virology, 185:120-131; Tsarev, S. A. et al. (1992), Proc. Natl. Acad. Sci. USA, 89:559-563; Fry, K. E. et al. (1992), Virus Genes, 6:173-185). Analysis of the DNA sequences have led investigators to hypothesize that the HEV genome is organized into three open reading frames (ORFs) and to hypothesize that these ORFs encode intact HEV proteins.
A partial DNA sequence of the genome of an HEV strain from Burma (Myanmar) is disclosed in Reyes et al., 1990, Science, 247:1335-1339. Tam et al., 1991, and Reyes et al., PCT Patent Application WO91/15603 published Oct. 17, 1991 disclose the complete nucleotide sequence and a deduced amino acid sequence of the Burma strain of HEV. These authors hypothesized that three forward open reading frames (ORFs) are contained within the sequence of this strain.
Ichikawa et al., 1991, Microbiol. Immunol., 35:535-543, discloses the isolation of a series of clones of 240-320 nucleotides in length upon the screening of a .lambda.gt11 expression library with sera from HEV-infected cynomolgus monkeys. The recombinant protein expressed by one clone was expressed in E. coli. This fusion protein is encoded by the 3' region of ORF-2 of the Myanmar strain of HEV.
The expression of additional proteins encoded within the 3' region of ORF-2 of a Mexican strain of HEV and of a Burmese strain of HEV is described in Yarbough et al., 1991 J. Virology, 65:5790-5797. This article describes the isolation of two cDNA clones derived from HEV. These clones encode the proteins in the 3' region of ORF-2. The clones were expressed in E. coli as fusion proteins.
Purdy et al., 1992, Archives of Virology, 123:335-349, and Favorov et al., 1992, J. of Medical Virology, 36:246-250, disclose the expression of a larger ORF-2 protein fragment from the Burma strain in E. coli. These references, as well as those previously discussed, only disclose the expression of a portion of the ORF-2 gene using bacterial expression systems. Successful expression of the full-length ORF-2 protein has not been disclosed until the present invention.
Comparison of the genome organization and morphological structure of HEV is most closely related to the caliciviruses. Of interest, the structural proteins of caliciviruses are encoded by the 3' portion of their genome (Neil, J. d. et al. (1991) J. Virol., 65:5440-5447; and Carter, M. J. et al. (1992), J. Arch. Virol., 122:223-235) and although there is no direct evidence that the 3' terminal part of the HEV genome also encodes the structural proteins, expression of certain small portions of the 3' genome region in bacterial cells resulted in production of proteins reactive with anti-HEV sera in ELISA and Western blots (Yarborough, et al., (1991); Ichikawa et al. (1991); Favorov et al. (1992) and Dawson, G. J. et al. (1992) J. Virol Meth; 38:175-186). However, the function of ORF-2 protein as a structural protein was not proven until the present invention.
The small proteins encoded by a portion of the ORF-2 gene have been used in immunoassay to detect antibodies to HEV in animal sera. The use of small bacterially expressed proteins as antigens in serological immunoassays has several potential drawbacks, first, the expression of these small proteins in bacterial cells of results in solubility problems and in non-specific cross-reactivity of patients' sera with E. coli proteins when crude E. coli lysates are used as antigens in immunoassays (Purdy et al. (1992)). Second, the use of Western blots as a first-line serological test for anti-HEV antibodies in routine epidemiology is impractical due to time and cost constraints. An ELISA using small-peptides derived from the 3'-terminal part of the HEV genome resulted in the detection of only 41% positives from known HEV-infected patients. Third, it has been shown that for many viruses, including Picornaviridae, important antigenic and immunogenic epitopes are highly conformation (Lemon, S. M. et al. (1991), in Hollinger, F. B., Lemon, S. M., Margolis, H. S. (eds): "Viral Hepatitis and Liver disease", Williams and Wilkins, Baltimore, 20-24). For this reason, it is believed that expression in a eukaryotic system of a complete ORF encoding an intact HEV gene would result in production of a protein which could form HEV-virus-like particles. Such a complete ORF protein would have an immunological structure closer to that of native capsid protein(s) than would the above-noted smaller proteins which represent only portions of the structural proteins of HEV. Therefore, these complete ORF proteins would likely serve as a more representative antigen and a more efficient immunogen than the currently-used smaller proteins.